mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Glauber Costa <glommer@redhat.com>
To: kvm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, avi@redhat.com,
	"H. Peter Anvin" <hpa@zytor.com>
Subject: [PATCH] deal with interrupt shadow state for emulated instruction
Date: Wed,  8 Apr 2009 17:42:46 -0400	[thread overview]
Message-ID: <1239226966-8205-1-git-send-email-glommer@redhat.com> (raw)

we currently unblock shadow interrupt state when we skip an instruction,
but failing to do so when we actually emulate one. This blocks interrupts
in key instruction blocks, in particular sti; hlt; sequences

If the instruction emulated is an sti, we have to block shadow interrupts.
The same goes for mov ss. pop ss also needs it, but we don't currently
emulate it. For sequences of two or more instructions of the same type
among those instructions, only the first one has this effect.

Without this patch, I cannot boot gpxe option roms at vmx machines.
This is described at https://bugzilla.redhat.com/show_bug.cgi?id=494469

Signed-off-by: Glauber Costa <glommer@redhat.com>
CC: H. Peter Anvin <hpa@zytor.com>
CC: Avi Kivity <avi@redhat.com>
---
 arch/x86/include/asm/kvm_host.h        |    2 +
 arch/x86/include/asm/kvm_x86_emulate.h |    2 +-
 arch/x86/kvm/svm.c                     |   21 +++++++++++++++-
 arch/x86/kvm/vmx.c                     |   39 ++++++++++++++++++++++++-------
 arch/x86/kvm/x86.c                     |   10 ++++++-
 arch/x86/kvm/x86_emulate.c             |   16 ++++++++++++-
 6 files changed, 75 insertions(+), 15 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 3fc4623..9853aa9 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -513,6 +513,8 @@ struct kvm_x86_ops {
 	void (*run)(struct kvm_vcpu *vcpu, struct kvm_run *run);
 	int (*handle_exit)(struct kvm_run *run, struct kvm_vcpu *vcpu);
 	void (*skip_emulated_instruction)(struct kvm_vcpu *vcpu);
+	void (*block_interrupt_shadow)(struct kvm_vcpu *vcpu);
+	void (*unblock_interrupt_shadow)(struct kvm_vcpu *vcpu);
 	void (*patch_hypercall)(struct kvm_vcpu *vcpu,
 				unsigned char *hypercall_addr);
 	int (*get_irq)(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/include/asm/kvm_x86_emulate.h b/arch/x86/include/asm/kvm_x86_emulate.h
index 6a15973..800c5b1 100644
--- a/arch/x86/include/asm/kvm_x86_emulate.h
+++ b/arch/x86/include/asm/kvm_x86_emulate.h
@@ -176,6 +176,6 @@ struct x86_emulate_ctxt {
 int x86_decode_insn(struct x86_emulate_ctxt *ctxt,
 		    struct x86_emulate_ops *ops);
 int x86_emulate_insn(struct x86_emulate_ctxt *ctxt,
-		     struct x86_emulate_ops *ops);
+		     struct x86_emulate_ops *ops, int *interruptibility);
 
 #endif /* _ASM_X86_KVM_X86_EMULATE_H */
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 3ffb695..86038a3 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -210,6 +210,22 @@ static int is_external_interrupt(u32 info)
 	return info == (SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_INTR);
 }
 
+static void svm_block_interrupt_shadow(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	svm->vmcb->control.int_state |= SVM_INTERRUPT_SHADOW_MASK;
+	vcpu->arch.interrupt_window_open = 0;
+}
+
+static void svm_unblock_interrupt_shadow(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK;
+	vcpu->arch.interrupt_window_open = (svm->vcpu.arch.hflags & HF_GIF_MASK);
+}
+
 static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
@@ -223,9 +239,8 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
 		       __func__, kvm_rip_read(vcpu), svm->next_rip);
 
 	kvm_rip_write(vcpu, svm->next_rip);
-	svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK;
 
-	vcpu->arch.interrupt_window_open = (svm->vcpu.arch.hflags & HF_GIF_MASK);
+	svm_unblock_interrupt_shadow(vcpu);
 }
 
 static int has_svm(void)
@@ -2660,6 +2675,8 @@ static struct kvm_x86_ops svm_x86_ops = {
 	.run = svm_vcpu_run,
 	.handle_exit = handle_exit,
 	.skip_emulated_instruction = skip_emulated_instruction,
+	.block_interrupt_shadow = svm_block_interrupt_shadow,
+	.unblock_interrupt_shadow = svm_unblock_interrupt_shadow,
 	.patch_hypercall = svm_patch_hypercall,
 	.get_irq = svm_get_irq,
 	.set_irq = svm_set_irq,
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index c6997c0..5158c2b 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -736,26 +736,45 @@ static void vmx_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
 	vmcs_writel(GUEST_RFLAGS, rflags);
 }
 
-static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
+static void vmx_block_interrupt_shadow(struct kvm_vcpu *vcpu)
 {
-	unsigned long rip;
-	u32 interruptibility;
+	u32 interruptibility = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO);
+	u32 interruptibility_mask = ((GUEST_INTR_STATE_STI | GUEST_INTR_STATE_MOV_SS));
 
-	rip = kvm_rip_read(vcpu);
-	rip += vmcs_read32(VM_EXIT_INSTRUCTION_LEN);
-	kvm_rip_write(vcpu, rip);
+	if (!(interruptibility & interruptibility_mask))
+		vmcs_write32(GUEST_INTERRUPTIBILITY_INFO,
+			     interruptibility | interruptibility_mask);
+	vcpu->arch.interrupt_window_open = 0;
+}
 
+static void vmx_unblock_interrupt_shadow(struct kvm_vcpu *vcpu)
+{
 	/*
 	 * We emulated an instruction, so temporary interrupt blocking
 	 * should be removed, if set.
 	 */
-	interruptibility = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO);
-	if (interruptibility & 3)
+	u32 interruptibility = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO);
+	u32 interruptibility_mask = ((GUEST_INTR_STATE_STI | GUEST_INTR_STATE_MOV_SS));
+
+	if (interruptibility & interruptibility_mask)
 		vmcs_write32(GUEST_INTERRUPTIBILITY_INFO,
-			     interruptibility & ~3);
+			     interruptibility & ~interruptibility_mask);
 	vcpu->arch.interrupt_window_open = 1;
 }
 
+static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
+{
+	unsigned long rip;
+
+	rip = kvm_rip_read(vcpu);
+	rip += vmcs_read32(VM_EXIT_INSTRUCTION_LEN);
+	kvm_rip_write(vcpu, rip);
+
+	/* skipping an emulated instruction also counts */
+	vmx_unblock_interrupt_shadow(vcpu);
+}
+
+
 static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
 				bool has_error_code, u32 error_code)
 {
@@ -3727,6 +3746,8 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.run = vmx_vcpu_run,
 	.handle_exit = vmx_handle_exit,
 	.skip_emulated_instruction = skip_emulated_instruction,
+	.block_interrupt_shadow = vmx_block_interrupt_shadow,
+	.unblock_interrupt_shadow = vmx_unblock_interrupt_shadow,
 	.patch_hypercall = vmx_patch_hypercall,
 	.get_irq = vmx_get_irq,
 	.set_irq = vmx_inject_irq,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 0bb4131..47daa23 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2364,7 +2364,7 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
 			u16 error_code,
 			int emulation_type)
 {
-	int r;
+	int r, interruptibility;
 	struct decode_cache *c;
 
 	kvm_clear_exception_queue(vcpu);
@@ -2412,7 +2412,13 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
 		}
 	}
 
-	r = x86_emulate_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
+	interruptibility = 0;
+	r = x86_emulate_insn(&vcpu->arch.emulate_ctxt, &emulate_ops, &interruptibility);
+
+	if (interruptibility)
+		kvm_x86_ops->block_interrupt_shadow(vcpu);
+	else
+		kvm_x86_ops->unblock_interrupt_shadow(vcpu);
 
 	if (vcpu->arch.pio.string)
 		return EMULATE_DO_MMIO;
diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c
index d7c9f6f..f3507ec 100644
--- a/arch/x86/kvm/x86_emulate.c
+++ b/arch/x86/kvm/x86_emulate.c
@@ -1350,7 +1350,8 @@ static inline int writeback(struct x86_emulate_ctxt *ctxt,
 }
 
 int
-x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
+x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops,
+						int *interruptibility)
 {
 	unsigned long memop = 0;
 	u64 msr_data;
@@ -1359,6 +1360,10 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
 	unsigned int port;
 	int io_dir_in;
 	int rc = 0;
+	static int movss_int_flag, movss_int_flag_old;
+
+	movss_int_flag_old = movss_int_flag;
+	movss_int_flag = 0;
 
 	/* Shadow copy of register state. Committed on successful emulation.
 	 * NOTE: we can copy them from vcpu as x86_decode_insn() doesn't
@@ -1610,6 +1615,13 @@ special_insn:
 
 		sel = c->src.val;
 		if (c->modrm_reg <= 5) {
+			if (c->modrm_reg == VCPU_SREG_SS) {
+				if (movss_int_flag_old)
+					*interruptibility = 1;
+				else
+					movss_int_flag = 1;
+			}
+
 			type_bits = (c->modrm_reg == 1) ? 9 : 1;
 			err = kvm_load_segment_descriptor(ctxt->vcpu, sel,
 							  type_bits, c->modrm_reg);
@@ -1864,6 +1876,8 @@ special_insn:
 		c->dst.type = OP_NONE;	/* Disable writeback. */
 		break;
 	case 0xfb: /* sti */
+		if (!(ctxt->eflags & X86_EFLAGS_IF))
+			*interruptibility = 1;
 		ctxt->eflags |= X86_EFLAGS_IF;
 		c->dst.type = OP_NONE;	/* Disable writeback. */
 		break;
-- 
1.5.6.6


             reply	other threads:[~2009-04-08 21:43 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-08 21:42 Glauber Costa [this message]
2009-04-11 16:37 ` Avi Kivity
2009-04-11 21:15   ` H. Peter Anvin
2009-04-13 20:06 Glauber Costa
2009-04-14  9:07 ` Gleb Natapov
2009-04-14  9:34 ` Avi Kivity
2009-04-14 16:07   ` H. Peter Anvin
2009-04-14 16:14     ` Avi Kivity
2009-04-14 16:25       ` H. Peter Anvin
2009-04-16  9:18         ` Avi Kivity
2009-04-16 22:40           ` H. Peter Anvin
2009-04-19  8:26             ` Avi Kivity
2009-04-14 17:31       ` Alan Cox
2009-04-14 17:32         ` H. Peter Anvin
2009-04-28 14:30 Glauber Costa
2009-04-30 11:50 ` Avi Kivity
2009-05-05 13:57   ` Glauber Costa
2009-05-05 18:40 Glauber Costa
2009-05-06  8:03 ` Gleb Natapov
2009-05-06 10:51   ` Avi Kivity
2009-05-08  5:25     ` Glauber Costa
2009-05-08  7:18       ` Gleb Natapov
2009-05-08 13:02         ` Glauber Costa

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1239226966-8205-1-git-send-email-glommer@redhat.com \
    --to=glommer@redhat.com \
    --cc=avi@redhat.com \
    --cc=hpa@zytor.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®